From www.bloodjournal.org by guest on September 2, 2017. For personal use only.

Review article

Interleukin-1 in the pathogenesis and treatment of inflammatory diseases

Charles A. Dinarello1,2

1Department of Medicine, University of Colorado, Aurora, CO; and 2Department of Medicine, Radboud University Nijmegen Medical Center, Nijmegen, The Netherlands

More than any other cytokine family, the family, IL-1␤ has emerged as a therapeu- eases such as gout, type 2 diabetes, heart IL-1 family of ligands and receptors is tic target for an expanding number of failure, recurrent pericarditis, rheumatoid primarily associated with acute and systemic and local inflammatory condi- arthritis, and smoldering myeloma also chronic inflammation. The cytosolic seg- tions called autoinflammatory diseases. are responsive to IL-1␤ neutralization. ment of each IL-1 receptor family member For these, neutralization of IL-1␤ results This review summarizes acute and contains the Toll-IL-1-receptor domain. in a rapid and sustained reduction in chronic inflammatory diseases that are This domain is also present in each Toll- disease severity. Treatment for autoim- treated by reducing IL-1␤ activity and like receptor, the receptors that respond mune diseases often includes immuno- proposes that disease severity is affected to microbial products and viruses. Since suppressive drugs whereas neutraliza- by the anti-inflammatory members of the Toll-IL-1-receptor domains are functional tion of IL-1␤ is mostly anti-inflammatory. IL-1 family of ligands and receptors. for both receptor families, responses to Although some autoinflammatory dis- (Blood. 2011;117(14):3720-3732) the IL-1 family are fundamental to innate eases are due to gain-of-function muta- immunity. Of the 11 members of the IL-1 tions for caspase-1 activity, common dis- Introduction

Since the 1996 publication in Blood of “Biologic Basis for IL-1 activity. Blocking IL-1, particularly IL-1␤, is now the Interleukin-1 in Disease,”1 there have been several major advances standard of therapy for a class of inflammatory syndromes termed in understanding a role for IL-1 in the pathogenesis of disease. “autoinflammatory” diseases (reviewed by Simon and van der Because of its property as a hematopoietic factor, IL-1 was Meer5 and Masters et al6). Autoinflammatory syndromes are administered to patients to improve recovery after BM transplanta- distinct from autoimmune diseases. In autoimmune diseases, the tion (human responses to IL-1 were reviewed in detail in 1996).1 T cell is associated with pathogenesis as the dysfunctional cell or Effective in reducing the duration of thrombocytopenia and “driver” of inflammation. Immunosuppressive therapies targeting leukopenia, recipients of IL-1 therapy experienced unacceptable T-cell function as well as antibodies that deplete T and B cells are signs and symptoms of systemic inflammation, including hypoten- effective in treating autoimmune diseases. In contrast, in auto- sion. Therefore, attention was initially focused on blocking IL-1 inflammatory diseases, the monocyte-macrophage is the dysfunc- activity in sepsis with the use of the naturally occurring IL-1 tional cell, which directly promotes inflammation. Autoinflamma- receptor antagonist (IL-1Ra), now known by its generic name tory conditions are characterized by recurrent bouts of fever with anakinra. There were 3 controlled trials of anakinra in human sepsis. debilitating local and systemic inflammation; they are often Although in each trial there was a reduction in 28-day all-causes responsive to IL-1␤ blockade (Table 1). In general, these diseases mortality compared with placebo-treated patients, the reductions did not are poorly controlled with immunosuppressive therapies, and reach statistical significance.2 The failure of blocking IL-1 to signifi- responses to blocking TNF␣, if any, are modest. In this review, a cantly reduce mortality in septic shock is not unusual, because most growing number of unrelated diseases often responsive to blocking ␤ anticytokines and anti-inflammatory agents have also failed in sepsis IL-1 are discussed. trials (reviewed in Eichacker et al3). Subsequently, attention focused on blocking IL-1 in noninfec- tious, chronic inflammatory conditions, such as rheumatoid arthri- ␣ ␤ tis. Anakinra is approved for reducing the signs and symptoms of From IL-1 and IL-1 to the IL-1 superfamily rheumatoid arthritis and slows the progressive joint destructive of ligands and receptors characteristics of the disease. Anakinra has also been administered Similarities in the IL-1 family to patients with smoldering/indolent myeloma at high risk of progression to multiple myeloma. In combination with a weekly Although the original IL-1 family comprised only IL-1␣ and low dose of dexamethasone, anakinra treatment provided a signifi- IL-1␤, the IL-1 family has expanded considerably. As shown in cant increase in the number of years of progression-free disease.4 Table 2, there are 11 members. The IL-1R family has also expanded Consistently, there are no organ toxicities or gastrointestinal to 9 distinct genes and includes coreceptors, decoy receptors, disturbances with anakinra or other parenteral therapies to reduce binding proteins, and inhibitory receptors.7 In terms of human

Submitted July 12, 2010; accepted January 18, 2011. Prepublished online as © 2011 by The American Society of Hematology Blood First Edition paper, February 8, 2011; DOI 10.1182/blood-2010-07-273417. The online version of this article contains a data supplement.

3720 BLOOD, 7 APRIL 2011 ⅐ VOLUME 117, NUMBER 14 From www.bloodjournal.org by guest on September 2, 2017. For personal use only.

BLOOD, 7 APRIL 2011 ⅐ VOLUME 117, NUMBER 14 IL-1 IN INFLAMMATORY DISEASES 3721

Table 1. Blocking IL-1␤ in treatment of acute and chronic signal is initiated (Figure 1; Table 2). Family members IL-36␣, ␤, inflammatory diseases and ␥ (formerly termed IL-1F6, IL-1F8, and IL-1F9) bind to the Classic autoinflammatory diseases IL-1Rp2, recruit the same IL-1RAcP, and a proinflammatory signal Familial Mediterranean fever (FMF) is initiated. IL-36Ra (formerly IL-1F5) also binds to the IL-1Rp2, Pyogenic arthritis, pyoderma gangrenosum, acne (PAPA)*† but does not recruit IL-1RAcP and acts as the natural-occurring Cryopyrin-associated periodic syndromes (CAPS) receptor antagonist for IL-36␣, ␤, and ␥. IL-33 binds to ST2, also Hyper IgD syndrome (HIDS) recruits the IL-1RAcP, but Th2-like properties characterize IL-33. Adult and juvenile Still disease Schnitzler syndrome Thus, the 6 proinflammatory members of the IL-1 family each TNF receptor-associated periodic syndrome (TRAPS) recruit the IL-1RAcP coreceptor with the TIR domain and MyD88 Blau syndrome; Sweet syndrome docks to each. Deficiency in IL-1 receptor antagonist (DIRA) Active precursors Probable autoinflammatory diseases Recurrent idiopathic pericarditis With the sole exception of IL-1Ra, each member of the IL-1 family Macrophage activation syndrome (MAS) is first synthesized as a precursor without a clear signal peptide for Urticarial vasculitis ␣ Antisynthetase syndrome processing and secretion, and none are found in the Golgi. IL-1 Relapsing chondritis and IL-33 are similar in that their precursor forms can bind to their Behçet disease respective receptor and trigger signal transduction. Although Erdheim-Chester syndrome (histiocytosis) recombinant mature forms of IL-1␣ and IL-33 are active, it remains Synovitis, acne, pustulosis, hyperostosis, osteitis (SAPHO) unclear whether the mature forms are representative of the Common diseases mediated by IL-1␤ naturally occurring IL-1␣ and IL-33 produced in vivo. The Rheumatoid arthritis‡ importance of the biologic activity of the IL-1␣ precursor is Periodic fever, aphthous stomatitis, pharyngitis, adenitis syndrome (PFAPA) discussed below. The precursor forms of IL-18 and IL-1␤ do not Urate crystal arthritis (gout) bind their respective receptors, are not active, and require cleavage Type 2 diabetes by either intracellular caspase-1 or extracellular neutrophilic pro- Smoldering multiple myeloma Postmyocardial infarction heart failure teases (see below). Osteoarthritis Dual-function cytokines Complete list of citations is in supplemental References (available on the Blood Web site; see the Supplemental Materials link at the top of the online article). IL-1␣ and IL-33 are “dual-function” cytokines in that in addition to *Incomplete responses to IL-1␤ blockade have been reported. binding to their respective cell surface receptors, the intracellular †The combination of IL-1 plus TNF␣ have been used in some disorders. precursor forms translocate to the nucleus and influence transcrip- ‡Rheumatoid arthritis is also treatable with other anticytokines, antireceptors, 9,10 ␣ immunomodulating agents, and B cell–depleting antibodies. tion. IL-33 is a bona fide transcription factor, and the IL-1 N-terminal amino acids contain a nuclear localization site (re- 7 ␣ disease, the properties of IL-1 itself still remain the model for viewed in Dinarello ). In general, the nuclear function of IL-1 or mediating inflammation. As shown in Figure 1, IL-1␣ or IL-1␤ IL-33 is transcription of proinflammatory genes. For example, ␣ bind first to the ligand-binding chain, termed type I (IL-1RI). This expression of the N-terminal amino acids of IL-1 stimulates IL-8 is followed by recruitment of the coreceptor chain, termed the production in the presence of complete blockade of the IL-1RI on 10 accessory protein (IL-1RAcP). A complex is formed of IL-1RI plus the cell surface. IL-37, formerly IL-1F7, also translocates to the 11 IL-1 plus the coreceptor. The signal is initiated with recruitment of nucleus. but, in the case of IL-37, there is a net decrease in 11,12 the adaptor protein MyD88 to the Toll-IL-1 receptor (TIR) domain inflammation. As such, IL-37 is a unique ligand in the IL-1 (see “IL-1 and TLR share similar functions”). Several kinases are family because it functions as an anti-inflammatory cytokine phosphorylated, NF-␬B translocates to the nucleus, and the expres- (see below). sion of a large portfolio of inflammatory genes takes place. Signal IL-1 and TLR share similar functions transduction in IL-1–stimulated cells has been reviewed in detail by Weber et al.8 The IL-1 family of receptors is unique in its class of Ig-like Similarly, IL-18 first binds to its ligand binding chain, IL-18R␣, receptors because of the presence of the TIR domain in cytoplasmic recruits its coreceptor the IL-18R␤ chain, and a proinflammatory segment of each member (reviewed in O’Neill13). The Toll protein

Table 2. The IL-1 family Family name Name Receptor Coreceptor Property

IL-1F1 IL-1␣ IL-1RI IL-1RacP Proinflammatory IL-1F2 IL-1␤ IL-1RI IL-1RAcP Proinflammatory IL-1F3 IL-1Ra IL-1RI NA Antagonist for IL-1␣, IL-1␤ IL-1F4 IL-18 IL-18R␣ IL-18R␤ Proinflammatory IL-1F5 IL-36Ra IL-1Rrp2 NA Antagonist for IL-36␣, IL-36␤, IL-36␥ IL-1F6 IL-36␣ IL-1Rrp2 IL-1RAcP Proinflammatory IL-1F7 IL-37 ?IL-18R␣* Unknown Anti-inflammatory IL-1F8 IL-36␤ IL-1Rrp2 IL-1RAcP Proinflammatory IL-1F9 IL-36␥ IL-1Rrp2 IL-1RAcP Proinflammatory IL-1F10 IL-38 Unknown Unknown Unknown IL-1F11 IL-33 ST2 IL-1RAcP Th2 responses, proinflammatory

NA indicates not applicable. *Question mark indicates needs confirmation. From www.bloodjournal.org by guest on September 2, 2017. For personal use only.

3722 DINARELLO BLOOD, 7 APRIL 2011 ⅐ VOLUME 117, NUMBER 14

Figure 1. Signaling and inhibition of signaling by IL-1Rs. (A) IL-1␣ (either precursor or mature) or mature IL-1␤ bind to the IL-1RI and with the IL-1RAcP forms the receptor heterodimeric complex. The TIR domain of each receptor chain approximate recruit MyD88, followed by phosphorylation of IL-1R–associated kinases (IRAKs) and inhibitor of NF␬B kinase ␤ (IKK␤), resulting in a signal to the nucleus. (B) In the brain and spinal cord, the variant IL-1RAcPb can form the heterodimeric complex with IL-1␣ or IL-1␤ and IL-1RI, but this complex fails to recruit MyD88, and there is inhibition of the IL-1 signal. The failure to recruit MyD88 may be because of an altered TIR domain (indicated as TIRb). (C) IL-1Ra binds to IL-1RI, but there is no signal because there is failure to form a complex with IL-1RAcP. (D) IL-1␤ binds to the IL-1RII, but lacking a cytoplasmic segment, there is no signal. (E) Because of an altered TIR domain (indicated as TIRb), SIGIRR inhibits IL-1 and TLR signaling. SIGIRR can form a complex with IL-33 (not shown) and inhibit IL-33 signaling. (F) IL-1Rrp2 binds IL-36␣, IL-36␤, or IL-36␥ and forms a complex with IL-1RAcP. The TIR domain of each receptor chain approximates and recruits MyD88 similar to that shown in panel A. (G) IL-36Ra binds to IL-1Rrp2 but fails to form a complex with IL-1RAcP. Thus, IL-36Ra prevents the binding of IL-36␣, IL-36␤, or IL-36␥ to IL-1Rrp2, and IL-36Ra is the natural receptor antagonist IL-36.

was originally studied for its role in embryogenesis of the fruit fly that different groups reported microbial products as ligands for Drosophila. After John Sims reported the cDNA of the IL-1RI, it TLRs. The TIR domain is perhaps one of the most conserved sequences was reported that the cytosolic segment of the IL-1RI shared a in both plants and animals.13 The near ubiquitous presence of the TIR 50-aa homologous domain with the Toll protein in Drosophila. The domain shows an unexpected duplication in evolution. Receptor signal- significance of this observation was unclear until a few months ing by the IL-1 family as well as for a large number of bacterial products later when Heguy et al14 demonstrated that the 50-aa domain in the and viral nucleic acids depend on the same TIR domain; therefore, the IL-1RI was essential for IL-1 activities. Was the same domain in inflammatory manifestations are similar. the Toll protein also functional for properties that were similar to Are endogenous mammalian proteins ligands for TLRs? IL-1 properties? It was not until 1996 that a role for the Toll protein in host defense of the fly was shown. But in 1988 van der Meer et There is considerable specificity of each IL-1 family ligand for its al15 had already reported that a low dose of IL-1 protected the host respective receptor. There appears to be less specificity in the TLRs against live infections. Thus, the link of the Toll protein to IL-1 and for the diversity of molecular structures in binding ligands, to nonspecific host defense against invading microorganisms had particularly for TLR2. For example, several products of Gram- already been made. In 1992, Charles Janeway replaced the term positive bacterial and mycobacterial organisms are ligands for “nonspecific resistance to infection” with “innate immune re- TLR2. However, there are reports showing a reduction in disease sponse.“ Still later, the existence a mammalian cell surface severity with blocking TLR2 in models that are IL-1 dependent but receptor for microbial products was shown and as reviewed by independent of microbial products.19,20 A role for TLRs in such O’Neill,13 these receptors are now called Toll-like receptors models raises the issue whether the extracellular segment of the (TLRs). All TLRs share the TIR domain with the IL-1RI. With TLRs recognizes a wide variety of endogenous proteins. Although the new term “innate immunity” came the term “acquired immu- there is some evidence for this possibility, particularly in TLR2, an nity,” and the concept that innate immunity affects acquired alternative explanation for disease modification in mice deficient in immunity. However, looking back, IL-1 is the best example of the specific TLRs may relate to a nonspecific coactivation or recruit- nonspecific nature of an “innate” cytokine. That innate immunity ment of the TIR domain of a particular TLR during signal affects acquired immunity had already been reported in 1979, with transduction of a specific cell surface or intracellular receptor. For the demonstration that human IL-1␤ enhanced antigen-specific example, during signal transduction after the binding of IL-1 to its responses of murine T cells.16 receptor, the TIR domain on TLR2 recruits MyD88 and contributes Humans have 11 TLRs.13 Each has a unique extracellular domain to the activation of NF␬B. This intracellular coactivation is often containing of leucine-rich repeats, which recognize a diverse number of termed “cross talk.” bacterial products, nucleic acids, and possibly some endogenous lipopro- teins. Intracellularly, however, each TLR has a functional TIR domain as does the IL-1RI. Similar to IL-1 receptors, TLRs recruit Limiting inflammation in the IL-1 family of the intracellular adaptor protein MyD88 to the TIR domain as the first step in signal transduction. The TIR domain is also present in 2 cytokines and receptors IL-1 family receptors that function to inhibit inflammation (Figure 1): Two receptor antagonists the single Ig IL-1–related receptor (SIGIRR)17 and the newly discovered IL-1R accessory protein, IL-1RAcPb.18 The TIR of these inhibitory As listed in Table 2, IL-1Ra binds to IL-1RI and is specific for receptors have amino acid substitutions, which alter charge distribution preventing the activity of IL-1␣ and IL-1␤. As shown in Figure 1, of the TIR domain and prevent the recruitment of MyD88.18 IL-36␣, IL-36␤, and IL-36␥ each bind to IL-1Rrp2, recruit Because deletion of the TIR domain from IL-1RI prevented IL-1RAcP, and initiate a proinflammatory signal. For example, IL-1 signaling,14 the presence of a functional TIR domain in both IL-36␣ inhibits the differentiation of adipocytes and induces IL-1 and TLRs suggested that microbial products such as endotoxin insulin resistance.21 However, IL-36Ra also binds to IL-1Rrp2, but could be ligands for TLRs. However, it was not until 1998-1999 similar to IL-1Ra, does not recruit IL-1RAcP. Therefore, IL-36Ra From www.bloodjournal.org by guest on September 2, 2017. For personal use only.

BLOOD, 7 APRIL 2011 ⅐ VOLUME 117, NUMBER 14 IL-1 IN INFLAMMATORY DISEASES 3723 is the specific receptor antagonist for IL-1Rrp2 and prevents the more severe responses to acute renal injury, which is because of a activity of IL-36␣, IL-36␤, and IL-36␥.22 In the skin, overexpres- greater infiltration of myeloid cells into the renal parenchyma.30 sion of IL-36␣ produces an inflammatory phenotype similar to that With the use of the same model of acute renal injury, caspase-1– of psoriasis and is antagonized by IL-36Ra.22 deficient mice are protected, most likely because of decreased IL-18 activity rather than decreased IL-1 effects.31 Thus, SIGIRR IL-1 receptor type II may keep IL-18 activity in the kidney at bay. IL-1 receptor type II (IL-1RII) does not signal because it lacks a IL-1RAcPb cytoplasmic domain, and, without a TIR domain, docking of MyD88 cannot take place (Figure 1). This receptor is expressed The IL-1RAcP serves as the coreceptor for IL-1␣, IL-1␤, IL-36␣, mostly on macrophages and B cells. IL-1RII binds IL-1␤ with a IL-36␤, IL-36␥, and IL-33, each a proinflammatory cytokine. greater affinity than that of IL-1RI, thereby sequestering IL-1␤. However, a unique form of the IL-1RAcP has been found in the Hence, cell surface IL-1RII functions as a decoy receptor for brain where it serves to decrease IL-1 activities.18 Termed IL- IL-1␤.23 Because IL-1RAcP is recruited to the IL-1RII–IL-1␤ 1RAcPb, this coreceptor forms the expected complex with IL-1 and complex, the decoy receptor also serves to sequester the accessory IL-1RI but does not recruit MyD88 or phosphorylate IL-1R– receptor from participating in IL-1 signaling from the IL-1RI. The associated kinase 4. Therefore, most IL-1 signaling is arrested. But, type II receptor is highly conserved and is found in bony fish where because some genes increase by the IL-1RI/IL-1RAcPb complex, ␤ it functions to inhibit inflammation because of IL-1 . The soluble partial IL-1 signaling takes place. Nevertheless, IL-1RAcPb func- ␤ (extracellular) domain binds both mature as well as the IL-1 tions as an inhibitory receptor chain only in the brain. Mice precursor. Although the soluble type II receptor is an effective deficient in IL-1RAcPb exhibit a normal inflammatory response in ␤ anti-inflammatory agent, for optimal neutralization of IL-1 , the the periphery but greater neurodegeneration in the brain. As such, soluble type II receptor, requires the soluble form of the IL- IL-1RAcPb could play a role in chronic inflammatory responses in 24 1RAcP. A splice variant of this receptor codes for the extracellular the brain by “buffering” IL-1–mediated neurodegeneration. domains and functions as an acute phase protein synthesized and Like SIGIRR, IL-1RAcPb contains the amino acid differences released from the liver. in its TIR domain that likely reduce binding of MyD88.18 In From a clinical perspective, expression of surface IL-1RII or addition to an altered TIR domain, IL-1RAcPb has a carboxyl circulating levels of soluble IL-1RII appear to affect the degree of extension of 140 aa. Carboxyl extensions are also present in inflammation in several diseases. Glucocorticoids induce expres- SIGIRR as well as 2 other members of the IL-1 receptor family, sion of surface IL-1RII, and this property may contribute to their TIGIRR-1 and TIGIRR-2. TIGIRR-2 is associated with an X- anti-inflammatory effects. For instance, in patients with an acute linked cognitive deficiency, which apparently is independent of loss in hearing because of an unknown autoimmune process, a IL-1 functions. Little is known about whether these C-terminal robust induction of IL-1RII by dexamethasone on blood monocytes segments contribute to the inhibitory properties of these receptors. correlated with clinical response to oral prednisone. In contrast, cells from patients unresponsive to prednisone did not increase A compelling case for the importance of the naturally expression of the decoy receptor.25 The production of IL-1␤ occurring IL-1Ra appears to play a role in endometriosis and related infertility. Peritoneal fluid from fertile and infertile women with endometrio- Rabbits passively immunized against their own endogenous IL- sis had lower soluble IL-1RII and higher IL-1␤ concentrations than 1Ra exhibit a worsening of colitis,32 and mice deficient in IL-1Ra healthy controls, and the increase in IL-1␤ was significant in develop spontaneous diseases such as a destructive arthritis,33 an women with endometriosis who reported pelvic pain.26 These and arteritis34 and a psoriatic-like skin eruption. Furthermore, mice several clinical studies support the concept that, like the balance of deficient in endogenous IL-1Ra develop aggressive tumors after IL-1 to IL-1Ra, the amount of surface or soluble IL-1RII affects exposure to carcinogens.35 These data support the concept that the disease severity. concentrations of IL-1␤ versus IL-1Ra affect the severity of some diseases. The single nucleotide polymorphism rs4251961 C allele SIGIRR is associated with lower circulating levels of IL-1Ra (reviewed in 36 37 37 The SIGIRR (also known as TIR817) is a unique receptor in the Dinarello and Larsen et al ) and is common in type 2 diabetes. IL-1 receptor family and functions to inhibit TLR- as well as The same polymorphism is associated with reduced survival in cytokine-mediated signals.27,28 As shown in Figure 1, the extracel- patients with colon carcinoma compared with those with the lular domain contains a single IgG-like domain; the cytosolic wild-type allele. The concept of an imbalance between IL-1␤ and portion does contain the TIR domain, but the 2 aa required for a IL-1Ra gained considerable legitimacy with reports of infants born functional TIR domain are absent. In addition, the C-terminal with nonfunctional IL-1Ra.38,39 Soon after birth, the affected contains an extra 95-aa extension, which may contribute to the infants exhibited impressive systemic and local inflammation. anti-inflammatory properties of this receptor.17 Although a ligand Multiple neutrophil-laden pustular skin eruptions, vasculitis, bone for SIGIRR has not been identified, the single Ig-like domain is abnormalities with large numbers of osteoclasts, osteolytic lesions, required for its inhibitory function. SIGIRR also forms a complex and sterile osteomyelitis were observed. The inflammation re- with the precursor form of IL-33,29 which may explain some of the sembled infection with sepsis-like multiorgan failure, but all anti-inflammatory properties of IL-33. For example, IL-33 reduces cultures were sterile. Treatment with anakinra was lifesaving; the the development of atherosclerosis as well as cardiomyocyte inflammation abated, and the bone lesions disappeared. IL-17 was hypertrophy. Intracellularly, SIGIRR reduces NF-␬B activation by prominently expressed in cells from these patients (see below). members of the IL-1 family (IL-1, IL-18, and IL-33), as well as by These findings are an extreme example that, without functional the family of TLRs. Compared with wild-type mice, mice deficient IL-1Ra, the activity of endogenous IL-1 is “unopposed” and in SIGIRR exhibit a severe inflammatory response to colitis28 and IL-1–driven inflammation can run rampant. From www.bloodjournal.org by guest on September 2, 2017. For personal use only.

3724 DINARELLO BLOOD, 7 APRIL 2011 ⅐ VOLUME 117, NUMBER 14

One can conclude that low levels of IL-1␤ can induce inflamma- complex systemic inflammatory diseases are driven by autoinflam- tion, but the presence of natural IL-1Ra is sufficient to limit the matory mechanisms, but adaptive pathways also seem to play a inflammation. IL-1Ra is found in the circulation of healthy subjects role. For example, in Behc¸et disease, there is a strong association in the range of 100-300 ng/mL, whereas IL-1␤ is in the picogram with HLA-B51, which probably contributes an autoimmune compo- per milliliter range and is not easily detectable by standard ELISA nent of the disease. methods in the same persons. In healthy humans, daily total Familial Mediterranean fever, familial cold-induced autoinflam- ␤ constitutive production of IL-1 has been calculated at 6 ng; in matory syndrome, and CAPS are classic examples of autoinflamma- patients with cryopyrin-associated periodic syndromes (CAPSs), a tory diseases.6,44 TNF receptor associated periodic syndrome was ␤ rare autoinflammatory disease associated with increased IL-1 the first disease to be labeled as autoinflammatory. As shown in ␤ 40 secretion, the calculated amount of IL-1 was 31 ng/d. These low Table 1, there are a growing number of diseases that fall under the ␤ amounts are consistent with reports of increased release of IL-1 category of being primarily autoinflammatory. One of the most from blood monocytes of patients with autoinflammatory diseases. consistent characteristics of these autoinflammatory diseases is a The increase is usually 5- to 10-fold more than that from blood rapid and sustained resolution of disease severity on treatment with monocytes of healthy subjects.41-43 IL-1␤ blockade, although there are exceptions. In the case of IL-37, a fundamental inhibitor of innate immune responses systemic onset juvenile idiopathic arthritis, there are often but not always dramatic responses to IL-1␤ blockade.41,45 Although antibod- The function of IL-37 (formerly IL-1 family member 7) has ies to the IL-6 receptor are also effective in this disease, a reduction remained elusive. Studies with the recombinant form of the in IL-1␤ activity probably reduces IL-6 production. Autoinflamma- cytokine have been inconsistent because of aggregation, although tory diseases are weakly responsive to TNF␣ neutralization, and, in some reports show that recombinant IL-37 binds to the IL-18R␣ some reports, disease worsens with TNF␣ blockade. Although chain. Nevertheless, a role for natural-occurring IL-37 in inflamma- autoimmune diseases respond to anti-TNF␣, CTLA4-Ig, anti–IL-6 tion was shown by suppressing endogenous IL-37 in human blood monocytes. In those studies, both lipopolysaccharide (LPS)– and receptor, depletion of CD20 B cells, depletion of CD3 T cells, IL-1␤–induced cytokine production increased significantly.12 Con- anti–IL-17, or anti–IL-12/IL-23 antibodies, these agents are often sistent with this observation, overexpression of IL-37 in macro- ineffective in patients with autoinflammatory diseases. For ex- phages or epithelial cells nearly completely suppressed production ample, in the case of Schnitzler syndrome with a monoclonal of proinflammatory cytokines. Mice with transgenic expression of gammopathy, depletion of B cells resulted in a dramatic reduction IL-37 were protected from LPS-induced shock and showed mark- in the paraprotein but was of no benefit on systemic symptoms. In edly improved lung and kidney functions with reduced liver contrast, anakinra resulted in a rapid improvement sometimes damage.12 IL-37 transgenic mice also exhibited lower concentra- within hours and a complete remission in disease activity within tions of circulating and tissue cytokines (72%-95% less) than days.46 Table 1 lists autoinflammatory diseases that, for the most wild-type mice and showed less dendritic cell activation. Previous part, respond to blocking IL-1 activity. studies have shown that IL-37 translocates to the nucleus via a caspase-1–dependent fashion.11 IL-37 interacts intracellularly with Clinical characteristics of systemic autoinflammatory diseases Sma- and Mad-related protein 3 (Smad3), and IL-37–expressing cells and transgenic mice showed less cytokine suppression when Systemic autoinflammatory diseases are syndromes.5,6,47 As listed endogenous Smad3 was depleted.12 It appears that IL-37 colocal- in Table 1, although many are rare, the clinical manifestations are izes to the nucleus with Smad3 and suppresses transcription of common. Fever, neutrophila, and elevated hepatic acute phase inflammatory genes induced by microbial products as well as proteins are characteristically present with painful joint, myalgias, cytokines themselves. Thus, IL-37 emerges as a natural suppressor and debilitating fatigue. These manifestations are consistent with of innate inflammatory and immune responses. Whether disease the responses observed in humans after administration of intrave- severity in humans is affected by the level of IL-37 expression nous IL-1; fever, neutrophilia, and elevated IL-6 and adrenocortico- remains to be determined. tropic hormone levels were observed at doses of IL-1␣ or IL-1␤ as low as 1-3 ng/kg.1 Patients with CAPS can present with a spectrum of disease IL-1␤ and the spectrum of autoinflammatory severity. In the mildest form, familial cold-induced autoinflamma- diseases tory syndrome, the episodes are triggered by cold exposure.44 A more severe form is Muckle-Wells syndrome, in which the The distinction between autoinflammatory and autoimmune inflammation is persistent and apparently not related to cold but diseases associated with the development of hearing loss and amyloidosis.48 Nearly all autoimmune diseases have an inflammatory component. The most severe form of CAPS is neonatal-onset multisystem However, autoinflammatory diseases appear to be primarily inflam- inflammatory disease, which, in addition to systemic inflammation, matory, with systemic as well as local manifestations that are often is also associated with joint deformities and developmental disabili- 42 ␤ periodic rather than progressive. Autoimmune diseases are primar- ties in early childhood. Often, on blocking IL-1 , patients with ily caused by dysregulation of adaptive immune responses and CAPS and other autoinflammatory diseases experience a rapid and because of pathologic antibodies or autoreactive T cells. In con- sustained cessation of symptoms as well as reductions in biochemi- trast, some classic autoinflammatory diseases are caused by genetic cal, hematologic, and functional markers of their disease.42,49,50 defects in innate inflammatory pathways and usually are mani- Because treatment with the soluble IL-1RI rilonacept or a monoclo- fested early in life. Typically, autoinflammatory diseases have no nal anti–IL-1␤ antibody (canakinumab) are equally effective in associations with HLA or MHC class II haplotypes, and there is and treating autoinflammatory diseases,49,50 the culprit in these diseases absence of autoreactive T cells or autoantibodies. However, some is IL-1␤ and not IL-1␣. From www.bloodjournal.org by guest on September 2, 2017. For personal use only.

BLOOD, 7 APRIL 2011 ⅐ VOLUME 117, NUMBER 14 IL-1 IN INFLAMMATORY DISEASES 3725

Caspase-1–dependent and –independent processing of the IL-1␤ precursor into an active cytokine

Increased release of IL-1␤ in autoinflammatory diseases

Fundamental to many autoinflammatory diseases is an increased release of active IL-1␤. Caspase-1 is the intracellular cysteine protease that cleaves the N-terminal 116 aa from the IL-1␤ precursor, thus converting the inactive precursor to the active “mature” cytokine. Caspase-1 exists in tissue macrophages and dendritic cells as an inactive zymogen and requires conversion to an active enzyme by autocatalysis. However, in circulating human blood monocytes, caspase-1 is present in an active state.51 Caspase-1 is also constitutively active in highly metastatic human melanoma cells.52 In general, the release of active IL-1␤ from blood mono- cytes is tightly controlled with Ͻ 20% of the total IL-1␤ precursor ␤ Figure 2. Steps in the processing and release of IL-1 induced by IL-1. (1) Pri- being processed and released. Although the release of active IL-1 mary blood monocytes, tissue macrophages or dendritic cells are activated by from blood monocytes of healthy subjects takes place over several either mature IL-1␤ or the IL-1␣ precursor with the formation of the IL-1 receptor hours, the process can be accelerated by increased levels of complex heterodimer comprised of the IL-1RI with IL-1RAcP. (2) Approximation of the intracellular TIR domains. (3) Recruitment of MyD88 and phosphorylation of extracellular ATP, which triggers the P2X7 purinergic receptor IL-1R–associated kinases (IRAKs) and inhibitor of NF␬B kinase ␤ (IKK␤). (4) Tran- (Figure 2).53,54 scription of IL-1␤. (5) Translation into the IL-1␤ mRNA takes place on polysomes. IL-1␤ mRNA is not bound to actin microfilaments but rather intermediate filaments. Activation of caspase-1 (6) ATP released from the activated monocyte/macrophage accumulates extra- cellularly.51 (7) Activation of the P2X7 receptor by ATP. (8) Efflux of potassium As shown in Figure 2, after ATPbinding to the P2X7 receptor, there from the cell after ATP binding to P2X7 receptor. (9) Fall in intracellular levels of is a rapid exit of potassium from the cell, and intracellular potassium. (10) The fall in intracellular potassium levels triggers the assembly of the components of the caspase-1 inflammasome with the conversion of potassium falls. With inhibition of ATP binding or in cells deficient procaspase-1 to active caspase-1. (11) Caspase-1 is found in the secretory in the P2X7 receptor, the amount of secreted IL-1␤ is low or lysosome together with the IL-1␤ precursor and lysosomal enzymes.53 Active absent.55 The fall in intracellular potassium is thought to bring caspase-1 cleaves the IL-1␤ precursor in the secretory lysosome, generating the active, carboxyl-terminal mature IL-1␤. (12) An influx of calcium with an increase about the oligomerization of a highly specialized group of intracel- in intracellular calcium levels. The rise in intracellular calcium activates phospha- lular proteins termed the “inflammasome,” which convert pro- tidylcholine-specific phospholipase C and calcium-dependent phospholipase A. caspase-1 to an active enzyme.56 Active caspase-1 then cleaves the (13) The release of mature IL-1␤, the IL-1␤ precursor, and the contents of the IL-1␤ precursor in specialized secretory lysosomes or in the secretory lysosomes by exocytosis in the absence of cell death. (14) Processing of the IL-1␤ precursor in the cytosol. Rab39a, a member of the GTPase family, cytosol, followed by secretion of “mature” IL-1␤ (Figure 2). contributes to the secretion of by helping traffic IL-1␤ from the cytosol into a Blood monocytes from patients with autoinflammatory syn- vesicle compartment. Exocytosis is another mechanism described in mouse dromes release more processed IL-1␤ than cells from healthy macrophages. (15) Mature IL-1␤ exists the cells via loss in membrane integrity, associated with the release of lactic dehydrogenase or microvesicles. TRAF subjects and thus likely account for the inflammation in these indicates TNF receptor-associated factor. diseases.40-43 An increase in the secretion of active IL-1␤ is observed in monocytes from patients with a gain-of-function mutation in a gene originally called cold-induced autoinflammatory syndrome-1.57 The mutations in this gene result in a single amino only a 5-fold more IL-1␤ is produced each day in patients with acid change in the intracellular protein named cryopyrin, because CAPS compared with healthy controls.40 after the exposure to cold, the patients develop fever and other Secretion of IL-1␤ manifestations of systemic inflammation. Cryopyrin is now termed nucleotide-binding domain and leucine-rich repeat-containing pro- As shown in Figure 2, multiple mechanisms have been reported for tein 3 (NLRP3). the secretion of the processed mature IL-1␤; these include the loss The gene cold-induced autoinflammatory syndrome-1was first of membrane integrity, a requirement for phospholipase C and reported by Hoffman et al in 2001,57 and a historical perspective of secretory lysosomes,53 and the shedding of plasma membrane this important discovery has been published.7 In 2004, NLRP3 was microvesicles or multivesicular bodies containing exosomes. In reported to associate with procaspase-1 and other intracellular general, the release of active IL-1␤ takes place before there is proteins to form a complex, which was termed the “inflam- significant release of lactate dehydrogenase, although some report masome.”56 As shown in Figure 2, these components activate that the release of IL-1␤ parallels that of lactate dehydrogenase. procaspase-1, resulting in active caspase-1, followed by the Pyroptosis is a caspase-1–dependent mechanism for cell death with processing and secretion of active IL-1␤. However, approximately the release of IL-1␤58 and may account for the release of active one-half of the patients with classic symptoms and biochemical IL-1␤ in ischemic disease.58 markers of CAPS, familial Mediterranean fever, and other autoin- Because caspase-1 is present in an active form in freshly flammatory diseases do not have mutations, and the basis for obtained blood monocytes from healthy subjects,51 the rate- increased secretion of IL-1␤ remains unknown. As noted above, limiting step in the secretion of IL-1␤ is at the transcriptional and the increase in IL-1␤ secretion in monocytes from patients with translational levels. But in monocytes from patients with single autoinflammatory diseases is modest compared with that of mono- amino acid mutation in NLRP3, the release of active IL-1␤ occurs cytes from healthy subjects. Despite severe systemic inflammation, without a requirement for a rapid fall in the level of intracellular From www.bloodjournal.org by guest on September 2, 2017. For personal use only.

3726 DINARELLO BLOOD, 7 APRIL 2011 ⅐ VOLUME 117, NUMBER 14

Like murine cells deficient in gp91phox, humans with chronic granulomatous disease have a mutation in p47-phox, with defective NADPH activity and are unable to generate ROSs. Blood mono- cytes from these patients release IL-1␤ after stimulation in vitro.65,66 Although the well-known inhibitor of ROSs, diphenylene iodonium, reduces the secretion of IL-1␤ in all studies, diphe- nylene iodonium inhibits IL-1␤ and also TNF␣ gene expression rather than caspase-1 activity.65 Thus, it appears that ROSs may dampen the inflammasome, and this may explain why patients with chronic granulomatous disease, who are unable to generate ROSs, Figure 3. Non–caspase-1 extracellular processing of the IL-1␤ precursor. The 269-aa-long IL-1␤ precursor is shown with the caspase-1 site at aspartic acid (D) at have increased inflammation with granulomatous lesions and a position 116. Extracellular protease sites are indicated by their amino acid recognition form of colitis indistinguishable from that found in Crohn disease. sites. The probable proteinase-3 site was derived from combinatorial methods of the It has also been shown that ROS induces intracellular N-acetylcys- specific substrate. teine, a potent antioxidant, which facilitates the secretion of active IL-1␤.67 Although ROS generation from mitochondria may contrib- ute to caspase-1 activation, there is the conundrum that, regardless potassium.43 Although often studied with the use of endotoxin- of the source of ROS, the concept is not supported by decades of induced synthesis of the IL-1␤ precursor, inflammation in autoin- failed placebo-controlled antioxidant trials. flammatory diseases is sterile. Continued treatment with anakinra ␣ results in a progressive reduction in the secretion of IL-1␤ from Role of IL-1 in ischemia-induced inflammation 42,43 monocytes compared with secretion before treatment. A reduc- Cells of mesenchymal origin constitutively contain preformed 42 tion in steady-state caspase-1 gene expression also takes place. IL-1␣ precursor, for example, keratinocytes, endothelial cells, and Thus, IL-1 itself increases the synthesis of caspase-1 as well as its epithelial cells of the lung, kidney, and gastrointestinal tract. If ␤ 59 own IL-1 precursor and may account for the autoinflammation released from its intracellular pool in healthy cells, the IL-1␣ of these diseases. precursor is inflammatory. Thus, it was not unexpected that robust Caspase-1–independent processing of the IL-1␤ precursor into inflammatory responses were observed when extracts of cellular ␣ an active cytokine contents from healthy tissues containing the IL-1 precursor on injection into the peritoneal cavity of mice.68,69 Importantly, In the arthritic joint and after blunt trauma, hypoxia, hemorrhage, inflammation was shown to be MyD88 and IL-1RI dependent but or exposure to irritants, there is a brisk myeloid response into the TLR independent.68,69 Intracellular IL-1␣ is found diffusely distrib- affected tissues dominated by neutrophils. Although IL-1␤ often uted in the cytosol as well as in the nucleus where it associates with plays a pivotal role in these conditions, caspase-1 may not be chromatin.70 The N-terminal domain of the IL-1␣ precursor required. For example, IL-1␤ is required for irritant-induced contains a nuclear localization sequence, which actively partici- inflammation in muscle tissue but is caspase-1 independent.60 pates in transcription10,71 and affects the senescence of cells. Similarly, cartilage destruction in joints61 and urate crystal–induced As shown in Figure 4, during the hypoxia of an ischemic event, inflammation62 are also IL-1␤ dependent but caspase-1 indepen- there is loss of membrane integrity in dying cells with the release of dent. In many models of sterile inflammation, cell death and release intracellular contents, which include the IL-1␣ precursor.70 Subse- of the IL-1␤ precursor from resident macrophages or infiltrating quently, resident tissue macrophages respond to IL-1␣ and produce monocytes take place. With the infiltrating neutrophils, there is also IL-1␤ (Figure 4). Although ischemia-induced inflammation is often a role for neutrophil proteases. As shown in Figure 3, these initiated by the IL-1␣ precursor, it rapidly becomes caspase-1 and proteases include elastase, chymases, granzyme A, cathepsin G, IL-1␤ dependent.72 Adenovirus infection induces a neutrophil and proteinase-3. In particular, proteinase-3 cleaves the inactive infiltration that is independent of TLR9 and NLRP3,73 and the IL-1␤ precursor close to the caspase-1 site, resulting in active inflammation is because of IL-1␣ activation of the IL-1RI with a IL-1␤.60,61 Moreover, inhibition of neutrophil proteases prevented brisk production of chemokines.73 Cell death by a caspase-1– urate crystal–induced peritonitis.62 These findings question whether dependent process called pyroptosis may also take place in caspase-1 is responsible for IL-1␤–dependent inflammation in ischemic tissues with the release of cell contents.58 patients with gout or any inflammatory process in which there is a In contrast, the same mesenchymally derived cells undergo robust infiltration of neutrophils in the release of the IL-1␤ apoptosis as part of normal tissue renewal. During apoptosis, precursor into the extracellular space. intracellular IL-1␣ concentrates in dense nuclear foci, and chroma- tin binding is not reversed by histone deacetylases or affected by Role of reactive oxygen species on caspase-1–dependent inhibitors of ROSs.70 As shown in Figure 4, in apoptotic cells, secretion of IL-1␤ IL-1␣ is retained within the chromatin fraction and is not present in Inflammation and carcinogenesis are associated with the generation the cytoplasmic contents. As such, lysates of cells undergoing of reactive oxygen species (ROSs), and ROSs have been proposed apoptotic death are biologically inactive.70 Thus, nuclear traffick- to activate the caspase-1 inflammasome.63 Because IL-1␤ appears ing and differential release during necrosis versus apoptosis to play a role in carcinogenesis,35 it is an attractive hypothesis that (Figure 4) show that inflammation by IL-1␣ is tightly controlled. ROSs directly activate the inflammasome to induce release of IL-1␤, periodontitis, and bone loss IL-1␤ and trigger inflammation. However, in cells deficient in gp91phox ROSs, levels are low and IL-1␤ levels are high, and in Epidemiologic studies have examined polymorphisms in the IL-1 cells deficient in super oxide dismutase-1, ROS levels are high but gene cluster (IL-1␣, IL-1␤, and IL-1Ra) with increased periodonti- caspase-1 is suppressed and IL-1␤ levels are low.64 tis and tooth loss as well as with coronary heart disease (CHD) and From www.bloodjournal.org by guest on September 2, 2017. For personal use only.

BLOOD, 7 APRIL 2011 ⅐ VOLUME 117, NUMBER 14 IL-1 IN INFLAMMATORY DISEASES 3727

secretion of large amounts of IL-1␤ from human macrophages.74 IL-1␤ is a known player in bone loss in patients with rheumatoid arthritis, and the amino acid sequence of purified osteoclast- activating factor is identical to that of IL-1␤. Thus, the induction of IL-1␤ in the periodontal space by the leukotoxin probably explains alveolar bone tooth loss with periodontitis. Because the inflammation in periodontitis appears to be IL-1␤–mediated, IL-1␤ entering the circulation from the local inflammation in the periodontal space may affect the distant insulin-producing ␤ cell in the pancreatic islet and account for the association of periodontitis with type 2 diabetes.

IL-1 in B- and T-cell responses

IL-1␤ as an adjuvant

A role for IL-1␤ in antibody production has been repeatedly reported (reviewed in Kelk et al75). Not unexpectedly, the most widely used adjuvant, aluminum hydroxide (alum), induces IL-1␤ Figure 4. Role of IL-1␣ in necrosis versus apoptosis. (1 left and right) In healthy cells of mesenchymal origin, the IL-1␣ precursor is found diffusely in the via a caspase-1–dependent mechanism. Previous studies have cytoplasm but also in the nucleus where it binds to chromatin. (2 right) During shown that mice deficient in IL-1␤ do not produce anti–sheep normal cell turnover, an apoptotic signal drives cytoplasmic IL-1␣ into the nucleus red blood cell antibodies, a T-dependent response. In wild-type and is no longer a dynamic in the cell. The cell shrinks. (2 left) Cells exposed to ␤ hypoxia begin to die, and nuclear IL-1␣ moves out of the nucleus into the mice, IL-1 causes a marked increase in the expansion of naive cytoplasm. Taking on water, the cell swells as the necrotic process begins. (3 left) and memory CD4 T cells in response to antigen and particularly As the necrotic process continues, there is loss of membrane integrity, and when used with LPS as a costimulant.76 In fact, when LPS is ␣ cytoplasmic contents containing the IL-1 precursor leak out. (3 right) Tissue used as an adjuvant ϳ 55% of the antibody response is because macrophages take up the apoptotic cell into endocytotic vesicles. (4) In the 76 vesicles, the apoptotic cell is digested, and there is no inflammatory response of IL-1. There are also striking increases in serum IgE and from the macrophage. (5 left) The IL-1␣ precursor is released into the extracellu- IgG1 levels. The role of IL-1␤ on the expansion and differentia- lar compartment and binds to IL-1RI expressed on adjacent cells or to resident tion of CD4 T cells is independent of IL-6 or CD28 but rather tissue macrophages. The tissue macrophage responds with synthesis of the IL-1␤ precursor as well as increased in caspase-1. From step 3 left, ATP is also because of IL-17– and IL-4–producing cells. In human T cells, released on cell death and activates the P2X7 receptor for activation of Th17 polarization induced by Mycobacterium tuberculosis is caspase-1. (6) Caspase-1 is activated by the inflammasome, cleaves the IL-1␤ IL-1␤ dependent77 via dectin-1 and TLR4 stimulation. This precursor, and mature, active IL-1␤ is released. Alternatively, IL-1␤ is released via pyroptosis. (7) Once released, IL-1␤ induces chemokine production, resulting in a observation has relevance to models of experimental autoim- chemoattractant gradient. (8) As the endothelium of the microcirculation ex- mune encephalomyelitis (EAE) and collagen-induced arthritis, presses adhesion molecule, blood neutrophils adhere and cross into the ischemic both of which require the complete Freund adjuvant, a mixture area. (9) With the infiltration of myeloid cells (monocytes and neutrophils), there is expanded inflammation, which extends beyond the initial area of ischemia. of killed M tuberculosis and the irritant croton oil. These results indicate that IL-1␤ signaling in T cells induces durable primary and secondary CD4 responses, most probably because of the induction of IL-17. type 2 diabetes. In a study in Germany, Ͼ 1500 subjects with A role for IL-1 in the generation of Th17 polarization increased levels of glycosylated hemoglobin were found to have significant periodontal disease compared with normoglycemic Increasingly, clinical studies indicate that T-cell differentiation into subjects. Diabetics with IL-1␤ promoter genotype C/T or T/T IL-17–producing cells plays a major role in some autoimmune exhibited a more severe form of periodontitis than their wild-type diseases, and IL-1 appears to play a pivotal role in the polarization 78,79 ␤ IL-1␤ counterparts, suggesting that carrying these polymorphisms of T cells toward Th17. The role of IL-1 in the generation of increases the inflammation. In another study, mean alveolar bone Th17 cells is also consistent with the inflammation in mice levels were measured in patients with CHD. The same polymor- deficient in IL-1Ra.80 In mice deficient in IL-1Ra, the activity of phisms were associated with bone loss. An IL-1Ra polymorphism endogenous IL-1 is unopposed, and spontaneous rheumatoid was also associated with both CHD and the highest level of bone arthritis-like disease develops33 but not in mice also deficient in loss. In patients with the IL-1␣ϩ4845T polymorphism, there was a IL-17.81 In mice specifically deficient in endogenous IL-1Ra in statistically significant correlation with acute coronary syndrome myeloid cells, collagen-induced arthritis appears to be because and severe periodontitis. of unopposed IL-1 activity, which drives IL-17.82 IL-1– The mechanism of IL-1␤ production in periodontitis has been dependent IL-17 production may be relevant to EAE, the rodent extensively studied with the use of bacteria isolated from patients model for multiple sclerosis. Mice deficient in both IL-1␣ and with long-standing disease. The most persistent bacterial infection IL-1␤ do not develop EAE, which is thought to be a failure to in periodontitis is Actinobacillus actinomycetemcomitans, which mount a Th17 response. In mice deficient in caspase-1, EAE is infects the periodontal space in many throughout the world and is markedly attenuated. Thus, IL-1 rather than TNF␣ blockade the most common cause of tooth loss.74 These organisms produce a reduces disease severity in mice subjected to EAE; in fact, leukotoxin, which induces degranulation and lysis in human blocking TNF␣ worsens the outcome of the EAE model. In neutrophils. The leukotoxin also activates caspase-1 and the rheumatoid arthritis, depleting CD20-bearing B cells results in From www.bloodjournal.org by guest on September 2, 2017. For personal use only.

3728 DINARELLO BLOOD, 7 APRIL 2011 ⅐ VOLUME 117, NUMBER 14 decreased IL-17 production,83 which may be because of a dexamethasone-mediated apoptosis. However, anakinra added to reduction in B-cell IL-1␤. these cocultures significantly reduced IL-6 by nearly 90%, and the Although there is no dearth of reports that adding IL-1 to combination of anakinra plus dexamethasone induced myeloma cultures induces IL-17, an essential role for IL-1 was shown when cell death.4 mice deficient in the IL-1R fail to induce IL-17 on antigen Patients with smoldering or indolent myeloma were selected challenge.84 Moreover, IL-23 fails to sustain IL-17 in IL-1R– with the clinical objective of slowing or preventing progression to deficient T-cells and TNF␣ and IL-6 enhancement of IL-23– active disease. On the basis of in vitro data, 47 patients with induced IL-17 is also IL-1 dependent.84 In a subsequent study of smoldering/indolent myeloma at high risk of progression to ϩ EAE, ␥/␦ CD4 T cells were the source of IL-17 but independent multiple myeloma were treated with daily anakinra for 6 months. ϩ of TCR engagement. ␥/␦ CD4 T cells contain the retinoic During the 6 months, there were decreases in C-reactive protein acid–related orphan receptor ␥ transcription factor and thus a direct (CRP) in most but not all patients, which paralleled a decrease in induction of IL-17 by IL-1 is fundamental to the Th17 paradigm. the plasma cell–labeling index, a measure of myeloma cell Therefore, it is probable that there is a cascade of IL-1␤–induced proliferation in unfractionated BM cells. After 6 months of IL-23 as well as IL-1–induced IL-6 for Th17 differentiation. anakinra, a low dose of dexamethasone (20 mg/week) was In human T cells, IL-1␤–dependent Th17 differentiation is added. Of the 47 patients who received anakinra (25 with because of the intermediate production of prostaglandin E2 (PGE2) dexamethasone), progression-free disease was Ͼ 3 years and in 85 from macrophages. This observation is consistent with ability of 8 patients Ͼ 4 years.4 Compared with historical experience, the ␤ 7 IL-1 to induce cyclooxygenase 2 (reviewed in Dinarello ) and findings indicate a significant failure to progress to active PGE2 to induce IL-6. Thus, an innate inflammatory response that is disease and a modest fall in CRP-predicted responders who ␤ IL-1 driven appears to play a pivotal role in the outcome of EAE. continued with stable disease. Patients with a decrease in serum ␥ To maintain Th1-driven responses, IFN suppresses the differentia- CRP of Ն 15% after 6 months of anakinra monotherapy resulted tion into the Th17 phenotype. One possible mechanism by which in progression-free disease Ͼ 3 years compared with 6 months ␥ ␥ IFN suppresses Th17 is by the suppression of IL-1. IFN reduces in patients with less than a 15% fall during anakinra therapy the induction of IL-1 by IL-1 and also of IL-1–induced PGE2 (P Ͻ .002). Thus, an effective reduction in IL-1␤ activity with 1 ␥ (reviewed in Dinarello ). Thus, IFN suppression of IL-1–driven the use of CRP as the marker for IL-1␤–induced IL-6 halts PGE2 as well as IL-1 itself may explain the reduction in Th17 progression to active myeloma. With the use of anakinra in differentiation. Ͼ 200 000 patients with rheumatoid arthritis or autoinflamma- tory diseases, no opportunistic infections, including M tuberculosis reactivation, have been reported.

Blocking IL-1␤–mediated disease Angiogenesis

Therapeutic agents Angiogenesis plays a pathologic role in multiple myeloma. In vitro and in animal models, there are a growing number of reports on Initially, anakinra was used to treat several unrelated chronic ␤ 88-90 inflammatory diseases as listed in Table 1. Today, these diseases IL-1 as a key cytokine in angiogenesis. Compared with ␤ are also successfully treated with neutralization by human thalidomide or its analogues, blocking IL-1 is essentially free of anti–IL-1␤ monoclonal Abs. For example in type 2 diabetes, toxic side effects. There is an ongoing trial at the National Institutes either anakinra or monoclonal Abs to IL-1␤ improve glycemic of Health of anakinra as an antiangiogenic therapy in patients with ␤ control.37,86,87 At this writing, one anti–IL-1␤ antibody (canaki- cutaneous melanoma. Because blocking IL-1 reduces IL-6 as well ␤ numab) has been approved for treating CAPS,50 whereas others as the proangiogenic chemokine IL-8, the use of IL-1 –blocking are presently in clinical trials. Rilonacept is a construct of the strategies may result in therapy in high-risk patients with smoldering/ 2 extracellular chains of the IL-1R complex (IL-1RI plus indolent myeloma or metastatic melanoma. There is also a role for ␤ 91 IL-1RAcP) fused to the Fc segment of IgG and has also been IL-1 in the angiogenic process of macular degeneration, and approved for use in CAPS.49 anakinra treatment in rheumatoid arthritis reduces the vasculariza- tion of the pannus. Targeting IL-1␤ in smoldering/indolent myeloma Role for IL-1 in GVHD Monoclonal gammopathy of undetermined significance and smol- dering myeloma present a challenge to medicine as the population Reducing cytokine activity in the development of GVHD is based ages. Several years of research has focused on the role of IL-1␤ and on preclinical studies. Of the 3 phases of GVHD, immunization, IL-6 in the pathogenesis of multiple myeloma. Similar to mature proliferation, and targeting of tissue, IL-1 is least important in the B cells, the myeloma plasma cell produces IL-1␤. In the microenvi- primary phases of the disease, in part, because IL-1 is not dominant ronment of the BM, stromal cells respond to low concentrations in the Th1 paradigm of T-cell and natural killer cell activation IL-1␤ and release large amounts of IL-6, which in turn promote the compared with IL-2, IL-15, and IL-18. In a placebo-controlled trial, survival and expansion of the myeloma cells. Although IL-6 is an patients were treated with anakinra 4 days before and 10 days after essential growth factor for myeloma cells, Abs to IL-6 have not allogeneic stem cell transplantation added to standards of therapy. been effective in treating the disease. Lust et al4 reasoned that, in Moderate-to-severe GVHD developed in both anakinra- and pla- the indolent stages of multiple myeloma, blocking IL-1␤ would cebo-treated patients.92 One can conclude that during conditioning provide better reduction of IL-6 activity. BM cells from patients and 10 days after transplantation, IL-1 plays no significant role in with smoldering myeloma were cocultured with a myeloma cell the development of the disease. In contrast, in patients with line actively secreting IL-1␤. Although the addition of dexametha- full-blown GVHD, blocking IL-1 reduced the severity of the sone reduced stromal cell IL-6 production, the amount of IL-6 disease.93 With the use of a continuous intravenous infusion of remained sufficiently high enough to protect the plasma cell against anakinra, there was improvement in the clinical score in 16 of From www.bloodjournal.org by guest on September 2, 2017. For personal use only.

BLOOD, 7 APRIL 2011 ⅐ VOLUME 117, NUMBER 14 IL-1 IN INFLAMMATORY DISEASES 3729

17 treated patients. Moreover, a decrease in the steady state mRNA same glycemic control compared with baseline requirements.37 for TNF␣ in peripheral blood mononuclear cells correlated with This observation suggests that blocking IL-1␤ even for a short improvement (P ϭ .001).93 It is probable that there is role for IL-1 period of time restores the function of the ␤ cells or possibly in the inflammatory manifestations of GVHD rather than the allows for partial regeneration of ␤ cells. Several trials that immunologic development of the disease. blocked TNF␣ in type 2 diabetes have succeeded in reducing CRP but did not improve glycemic control. The findings of Treating gout with IL-1␤ neutralization anakinra therapy in patients with type 2 diabetes have been There is a long association of IL-1 production with urate crystals confirmed with canakinumab, as well as another neutralizing and a causal relationship of IL-1 with gout.94,95 Some patients monoclonal antibody to IL-1␤.87 The data also provide evidence with recurrent attacks of gouty arthritis resistant to colchicine and that short-term blockade of IL-1␤ restores the function of the ␤ other standards of therapy often require steroids to control disease cells or possible regeneration. flares. When treated with anakinra, rilonacept, or canakinumab, a The above studies are consistent with the concept that type2 rapid, sustained, and remarkable reduction in pain and objective diabetes is a chronic IL-1–mediated disease and could be signs of reduced inflammation have been observed.96-98 The effect classified as an autoinflammatory disease in which IL-1␤– of IL-1 blockade appears to be superior to that of steroids and mediated inflammation progressively destroys the insulin- result in prolonged periods without flares. Although some producing ␤ cells. The IL-1␤ can come from the ␤ cell itself, studies report that urate crystals directly activate the NLRP3 for from infiltrating macrophages into the islet, or from the adipose processing of the IL-1␤ precursor,99 it is probable that 2 signals tissue stores. A requirement for caspase-1 in regulating IL-1␤ are required; urate crystals plus free fatty acids account for production from the insulin-producing ␤ cell as well as the IL-1␤ activity in flares of gout.100,101 Given the characteristic adipocyte provides a molecular mechanism. Differentiation of neutrophilic infiltration in gouty joints, it is also probable that the adipocyte is caspase-1 dependent.108 In diabetic mice, ␤ the IL-1 precursor is processed extracellularly by neutrophilic administration of a caspase-1 inhibitor reduces insulin resis- enzymes as shown in Figure 3. tance; in mice deficient in caspase-1, there is improved sensitiv- 108 Effects of IL-1R blockade in stroke ity to insulin. Because type 2 diabetes increases risk of cardiovascular events, blocking IL-1␤ activity in these patients Although infection, particularly septic shock, evokes a “cytokine may also reduce the incidence in myocardial infarction and storm” as part of a systemic inflammatory response, in many stroke as reviewed in Dinarello.106 ischemic diseases, such as acute lung injury, thrombotic stroke, ␤ acute renal failure, myocardial infarction, and hepatic failure, IL-1 as a mediator of cartilage destruction in osteoarthritis inflammation is characterized by cell death and infiltration of In the joint, IL-1␤ is the mediator of reduced chondrocyte neutrophils into the ischemic area. The induction of IL-1␤ in proteoglycan synthesis, increased synthesis of matrix metalloprotei- human monocytes exposed to hypoxia was first reported in nases, and the release of nitric oxide.109 Mice deficient in IL-1␤ are 1991.102 Subsequently, many studies have shown an essential protected from inflammation-induced arthritis. The role of IL-1␤ in role for IL-1 and specifically for IL-1␤ and caspase-1 in various the destructive processes of osteoarthritis has also been studied in models of ischemic injury without and with reperfusion. A rabbits, pigs, dogs, and horses. In a placebo-controlled trial of randomized, double-blind, placebo-controlled trial of anakinra intra-articular anakinra in patients with knee osteoarthritis, there was carried out in patients with acute stroke. Three days of a was a clear dose-dependent (50 mg vs 150 mg) reduction in pain high dose of intravenous anakinra was administered, the same and stiffness scores, but the benefit did not extend beyond dose used in the sepsis trials.103 Peripheral white blood cells, 1 month.110 The modest reduction may be because of the heteroge- neutrophil counts, CRP, and IL-6 concentrations were lower in patients treated with anakinra than with placebo. After 3 months, neity of the osteoarthritis population in general but also to the short the anakinra-treated patients exhibited less loss of cognitive duration of IL-1RI blockade by anakinra. 103 function than the placebo group. Does IL-1 contribute to heart failure after myocardial infarction? ␤ Reducing IL-1 in type 2 diabetes ST elevation myocardial infarction has a high risk of death for For Ͼ 25 years, the cytotoxic effects of IL-1␤ for the insulin- patients, but patients who survive the acute event progress to producing pancreatic ␤ cell have been studied.104 Subsequently, it chronic heart failure because of a loss of viable myocardium. In a was shown that high concentrations of glucose stimulated IL-1␤ randomized, placebo-controlled trial of patients with ST elevation production from the ␤ cell itself,105 implicating a role for IL-1␤ in myocardial infarction, daily anakinra was added to standard of type 2 diabetes.106 In addition, free fatty acids act together with therapy the day after angioplasty and continued for 14 days. Serial glucose to stimulate IL-1␤ from the ␤ cell. The production of imaging and echocardiographic studies were performed over the IL-1␤ also increases the deposition of amyloid via activation of following 10-14 weeks.111 Left ventricular remodeling was signifi- NLRP3.107 In support of these studies, gene expression for cantly reduced with anakinra treatment and consistent with the IL-1␤ was Ͼ 100-fold higher in ␤ cells from patients with type 2 reductions in CRP compared with patients receiving 14 days of diabetes than from patients without diabetes. The clinical proof placebo. After 18 months, 60% of placebo-treated patients had of a role for IL-1␤ in the pathogenesis of type 2 diabetes came from developed stage IV heart failure, whereas there were none in the a randomized, placebo-controlled study of anakinra for 13 weeks, anakinra-treated patients. These findings are consistent with experi- in which improved insulin production and glycemic control mental myocardial infarction in mice in which blocking IL-1␤ associated with decreased CRP and IL-6 levels was reported.86 results is a reduction in postinfarction remodeling.72 It remains to In the 39 weeks after the 13-week course of anakinra, patients be tested whether chronic heart failure in non–postmyocardial who responded to anakinra used 66% less insulin to obtain the infarction patients is reduced by IL-1␤ blockade. From www.bloodjournal.org by guest on September 2, 2017. For personal use only.

3730 DINARELLO BLOOD, 7 APRIL 2011 ⅐ VOLUME 117, NUMBER 14

Acknowledgments Authorship I thank Antonio Abbate, Leo Joosten, Mihai Netea, Frank van de Veerdonk, and Jos van der Meer for many helpful suggestions in Contribution: C.A.D. was the single author of this paper. the preparation of this manuscript. I also thank Kenneth X. Probst Conflict-of-interest disclosure: The author declares no compet- for the professional drawings of the figures. ing financial interests. This work was supported by National Institutes of Health Correspondence: Charles A. Dinarello, University of Colorado (grants AI-15614, CA-04 6934) and Juvenile Diabetes Research Denver, 12700 E 19th Ave B168, Aurora, CO 80045; e-mail: Foundation (26-2008-893). [email protected]. References

1. Dinarello CA. Biological basis for interleukin-1 in an IL-1R/TLR family member with regulatory func- 32. Ferretti M, Casini-Raggi V, Pizarro TT, Eisenberg SP, disease. Blood. 1996;87(6):2095-2147. tions in inflammation and T cell polarization. Trends Nast CC, Cominelli F. Neutralization of endoge- 2. Opal SM, Fisher CJJ, Dhainaut JF, et al. Confir- Immunol. 2009;30(9):439-446. nous IL-1 receptor antagonist exacerbates and matory interleukin-1 receptor antagonist trial in 18. Smith DE, Lipsky BP, Russell C, et al. A central prolongs inflammation in rabbit immune colitis. severe sepsis: a phase III, randomized, double- nervous system-restricted isoform of the interleukin-1 J Clin Invest. 1994;94(1):449-453. blind, placebo-controlled, multicenter trial. Crit receptor accessory protein modulates neuronal re- 33. Horai R, Saijo S, Tanioka H, et al. Development of Care Med. 1997;25(7):1115-1124. sponses to interleukin-1. Immunity. 2009;30(6):817- chronic inflammatory arthropathy resembling 3. Eichacker PQ, Parent C, Kalil A, et al. Risk and 831. rheumatoid arthritis in interleukin 1 receptor an- the efficacy of antiinflammatory agents: retro- 19. O’Neill LA, Bryant CE, Doyle SL. Therapeutic tar- tagonist-deficient mice. J Exp Med. 2000;191(2): spective and confirmatory studies of sepsis. Am J geting of toll-like receptors for infectious and in- 313-320. Respir Crit Care Med. 2002;166(9):1197-1205. flammatory diseases and cancer. Pharmacol Rev. 34. Nicklin MJ, Hughes DE, Barton JL, Ure JM, Duff GW. 4. Lust JA, Lacy MQ, Zeldenrust SR, et al. Induction 2009;61(2):177-197. Arterial inflammation in mice lacking the interleukin 1 of a chronic disease state in patients with smol- 20. Arslan F, Smeets MB, O’Neill LA, et al. Myocar- receptor antagonist gene. J Exp Med. 2000;191(2): dering or indolent multiple myeloma by targeting dial ischemia/reperfusion injury is mediated by 303-312. interleukin 1beta-induced interleukin 6 production leukocytic toll-like receptor-2 and reduced by sys- 35. Krelin Y, Voronov E, Dotan S, et al. Interleukin- and the myeloma proliferative component. Mayo temic administration of a novel anti-toll-like 1beta-driven inflammation promotes the develop- Clin Proc. 2009;84(2):114-122. receptor-2 antibody. Circulation. 2010;121(1): ment and invasiveness of chemical carcinogen- 5. Simon A, van der Meer JW. Pathogenesis of fa- 80-90. induced tumors. Cancer Res. 2007;67(3):1062- milial periodic fever syndromes or hereditary au- 21. van Asseldonk EJ, Stienstra R, Koenen TB, et al. 1071. toinflammatory syndromes. Am J Physiol Regul The effect of the interleukin-1 cytokine family 36. Dinarello CA. Why not treat human cancer with Integr Comp Physiol. 2007;292(1):R86-98. members IL-1F6 and IL-1F8 on adipocyte differ- interleukin-1 blockade? Cancer Metastasis Rev. entiation. Obesity. 2010;18(11):2234-2236. 2010;29(2):317-329. 6. Masters SL, Simon A, Aksentijevich I, Kastner DL. Horror autoinflammaticus: the molecular pathophysi- 22. Blumberg H, Dinh H, Trueblood ES, et al. Oppos- 37. Larsen CM, Faulenbach M, Vaag A, Ehses JA, ology of autoinflammatory disease. Annu Rev Immu- ing activities of two novel members of the IL-1 Donath MY, Mandrup-Poulsen T. Sustained ef- nol. 2009;27:621-668. ligand family regulate skin inflammation. J Exp fects of interleukin-1 receptor antagonist treat- Med. 2007;204(11):2603-2614. ment in type 2 diabetes. Diabetes Care. 2009; 7. Dinarello CA. Immunological and inflammatory 32(9):1663-1668. functions of the interleukin-1 family Ann Rev Im- 23. Colotta F, Re F, Muzio M, et al. Interleukin-1 type munol. 2009;27:519-550. II receptor: a decoy target for IL-1 that is regu- 38. Aksentijevich I, Masters SL, Ferguson PJ, et al. lated by IL-4. Science. 1993;261(5120):472-475. An autoinflammatory disease with deficiency of 8. Weber A, Wasiliew P, Kracht M. Interleukin-1 the interleukin-1-receptor antagonist. N Engl (IL-1) pathway. Sci Signal. 2010;3:cm1. 24. Smeets RL, Joosten LA, Arntz OJ, et al. Soluble interleukin-1 receptor accessory protein amelio- J Med. 2009;360(23):2426-2437. 9. Carriere V, Roussel L, Ortega N, et al. IL-33, the rates collagen-induced arthritis by a different 39. Reddy S, Jia S, Geoffrey R, et al. An autoinflam- IL-1-like cytokine ligand for ST2 receptor, is a mode of action from that of interleukin-1 receptor matory disease due to homozygous deletion of chromatin-associated nuclear factor in vivo. Proc antagonist. Arthritis Rheum. 2005;52(7):2202- the IL1RN locus. N Engl J Med. 2009;360(23): Natl Acad Sci U S A. 2007;104(1):282-287. 2211. 2438-2444. 10. Werman A, Werman-Venkert R, White R, et al. 25. Vambutas A, DeVoti J, Goldofsky E, Gordon M, 40. Lachmann HJ, Lowe P, Felix SD, et al. In vivo The precursor form of IL-1alpha is an intracrine Lesser M, Bonagura V. Alternate splicing of regulation of interleukin 1beta in patients with proinflammatory activator of transcription. Proc interleukin-1 receptor type II (IL1R2) in vitro cor- cryopyrin-associated periodic syndromes. J Exp Natl Acad Sci U S A. 2004;101(8):2434-2439. relates with clinical glucocorticoid responsiveness Med. 2009;206(5):1029-1036. 11. Sharma S, Kulk N, Nold MF, et al. The IL-1 family in patients with AIED. PLoS One. 2009;4:e5293. 41. Pascual V, Allantaz F, Arce E, Punaro M, member 7b translocates to the nucleus and 26. Akoum A, Al-Akoum M, Lemay A, Maheux R, Banchereau J. Role of interleukin-1 (IL-1) in the down-regulates proinflammatory cytokines. J Im- Leboeuf M. Imbalance in the peritoneal levels of pathogenesis of systemic onset juvenile idio- munol. 2008;180(8):5477-5482. interleukin 1 and its decoy inhibitory receptor type pathic arthritis and clinical response to IL-1 block- 12. Nold MF, Nold-Petry CA, Zepp JA, Palmer BE, II in endometriosis women with infertility and pel- ade. J Exp Med. 2005;201(9):1479-1486. Bufler P, Dinarello CA. IL-37 is a fundamental in- vic pain. Fertil Steril. 2008;89(6):1618-1624. 42. Goldbach-Mansky R, Dailey NJ, Canna SW, et al. hibitor of innate immunity. Nat Immunol. 2010; 27. Wald D, Qin J, Zhao Z, et al. SIGIRR, a negative Neonatal-onset multisystem inflammatory dis- 11(11):1014-1022. regulator of Toll-like receptor-interleukin 1 recep- ease responsive to interleukin-1beta inhibition. 13. O’Neill LA. The interleukin-1 receptor/Toll-like re- tor signaling. Nat Immunol. 2003;4(9):920-927. N Engl J Med. 2006;355(6):581-592. ceptor superfamily: 10 years of progress. Immu- 28. Garlanda C, Riva F, Polentarutti N, et al. Intestinal 43. Gattorno M, Tassi S, Carta S, et al. Pattern of nol Rev. 2008;226:10-18. inflammation in mice deficient in Tir8, an inhibitory interleukin-1beta secretion in response to lipo- 14. Heguy A, Baldari C, Bush K, et al. Internalization member of the IL-1 receptor family. Proc Natl polysaccharide and ATP before and after and nuclear localization of interleukin 1 are not Acad Sci U S A. 2004;101(10):3522-3526. interleukin-1 blockade in patients with CIAS1 mu- sufficient for function. Cell Growth Differ. 1991; 29. Bulek K, Swaidani S, Qin J, et al. The essential tations. Arthritis Rheum. 2007;56(9):3138-3148. 2(7):311-315. role of single Ig IL-1 receptor-related molecule/ 44. Hoffman HM, Rosengren S, Boyle DL, et al. Pre- 15. van der Meer JWM, Barza M, Wolff SM, Toll IL-1R8 in regulation of Th2 immune re- vention of cold-associated acute inflammation in Dinarello CA. A low dose of recombinant inter- sponse. J Immunol. 2009;182(5):2601-2609. familial cold autoinflammatory syndrome by leukin 1 protects granulocytopenic mice from 30. Lech M, Avila-Ferrufino A, Allam R, et al. Resi- interleukin-1 receptor antagonist prevents. Lan- lethal gram-negative infection. Proc Natl Acad dent dendritic cells prevent postischemic acute cet. 2004;364(9447):1779-1785. Sci U S A. 1988;85(5):1620-1623. renal failure by help of single Ig IL-1 receptor- 45. Gattorno M, Piccini A, Lasiglie D, et al. The pat- 16. Rosenwasser LJ, Dinarello CA, Rosenthal AS. related protein. J Immunol. 2009;183(6):4109- tern of response to anti-interleukin-1 treatment Adherent cell function in murine T-lymphocyte 4118. distinguishes two subsets of patients with sys- antigen recognition, IV: enhancement of murine 31. Melnikov VY, Ecder T, Fantuzzi G, et al. Impaired temic-onset juvenile idiopathic arthritis. Arthritis T-cell antigen recognition by human leukocytic IL-18 processing protects caspase-1-deficient Rheum. 2008;58(5):1505-1515. pyrogen. J Exp Med. 1979;150(3):709-714. mice from ischemic acute renal failure. J Clin In- 46. Eiling E, Moller M, Kreiselmaier I, Brasch J, 17. Garlanda C, Anders HJ, Mantovani A. TIR8/SIGIRR: vest. 2001;107(9):1145-1152. Schwarz T. Schnitzler syndrome: treatment failure From www.bloodjournal.org by guest on September 2, 2017. For personal use only.

BLOOD, 7 APRIL 2011 ⅐ VOLUME 117, NUMBER 14 IL-1 IN INFLAMMATORY DISEASES 3731

to rituximab but response to anakinra. J Am Acad dismutase 1 regulates caspase-1 and endotoxic is required for the spontaneous development of Dermatol. 2007;57(2):361-364. shock. Nat Immunol. 2008;9(8):866-872. destructive arthritis in mice deficient in IL-1 recep- 47. Goldbach-Mansky R, Kastner DL. Autoinflamma- 65. van de Veerdonk FL, Smeekens SP, Joosten LA, tor antagonist. Proc Natl Acad Sci U S A. 2003; tion: the prominent role of IL-1 in monogenic au- et al. Reactive oxygen species-independent acti- 100(10):5986-5990. toinflammatory diseases and implications for vation of the IL-1beta inflammasome in cells from 82. Lamacchia C, Palmer G, Seemayer CA, common illnesses. J Allergy Clin Immunol. 2009; patients with chronic granulomatous disease. Talabot-Ayer D, Gabay C. Enhanced Th1 and 124(6):1141-1149; quiz 1150-1141. Proc Natl Acad Sci U S A. 2010;107(7):3030- Th17 responses and arthritis severity in mice with a deficiency of myeloid cell-specific 48. Hawkins PN, Lachmann HJ, Aganna E, 3033. interleukin-1 receptor antagonist. Arthritis McDermott MF. Spectrum of clinical features in 66. Meissner F, Seger RA, Moshous D, Fischer A, Rheum. 2010;62(2):452-462. Muckle-Wells syndrome and response to anak- Reichenbach J, Zychlinsky A. Inflammasome ac- inra. Arthr Rheumat. 2004;50(2):607-612. tivation in NADPH oxidase defective mononu- 83. van de Veerdonk FL, Lauweys B, DiPadova F, et al. The anti-CD20 antibody rituximab reduces 49. Hoffman HM, Throne ML, Amar NJ, et al. Efficacy clear phagocytes from patients with chronic the T helper response. Arthr Rheumat. In press. and safety of rilonacept (interleukin-1 trap) in pa- granulomatous disease. Blood. 2010;116(9): tients with cryopyrin-associated periodic syn- 1570-1573. 84. Sutton C, Brereton C, Keogh B, Mills KH, Lavelle EC. dromes: results from two sequential placebo- 67. Tassi S, Carta S, Delfino L, et al. Altered redox A crucial role for interleukin (IL)-1 in the induction of controlled studies. Arthritis Rheum. 2008;58(8): state of monocytes from cryopyrin-associated IL-17-producing T cells that mediate autoimmune 2443-2452. periodic syndromes causes accelerated IL-1beta encephalomyelitis. J Exp Med. 2006;203(7):1685- secretion. Proc Natl Acad Sci U S A. 2010; 1691. 50. Lachmann HJ, Kone-Paut I, Kuemmerle- 107(21):9789-9794. 85. Chizzolini C, Chicheportiche R, Alvarez M, et al. Deschner JB, et al. Use of canakinumab in the Prostaglandin E2 synergistically with interleukin-23 cryopyrin-associated periodic syndrome. 68. Chen CJ, Kono H, Golenbock D, Reed G, Akira S, favors human Th17 expansion. Blood. 2008;112(9): N Engl J Med. 2009;360(23):2416-2425. Rock KL. Identification of a key pathway required for the sterile inflammatory response triggered by dying 3696-3703. 51. Netea MG, Nold-Petry CA, Nold MF, et al. Differ- cells. Nat Med. 2007;13(7):851-856. 86. Larsen CM, Faulenbach M, Vaag A, et al. Interleukin- ential requirement for the activation of the inflam- 1-receptor antagonist in type 2 diabetes mellitus. masome for processing and release of IL-1beta in 69. Eigenbrod T, Park JH, Harder J, Iwakura Y, N Engl J Med. 2007;356(15):1517-1526. monocytes and macrophages. Blood. 2009; Nunez G. Cutting edge: critical role for mesothe- 113(10):2324-2335. lial cells in necrosis-induced inflammation through 87. Donath MY, Weder C, Brunner A, et al. XOMA the recognition of IL-1 alpha released from dying 052, a potential disease modifying anti-1L–1beta 52. Okamoto M, Liu W, Luo Y, et al. Constitutively cells. J Immunol. 2008;181(12):8194-8198. antibody, shows sustained HbA1c reductions active inflammasome in human melanoma cells 3 months after a single injection with no in- 70. Cohen I, Rider P, Carmi Y, et al. Differential re- mediating autoinflammation via caspase-1 pro- creases in safety parameters in subjects with lease of chromatin-bound IL-1alpha discriminates cessing and secretion of interleukin-1beta. J Biol Type 2 diabetes. Diabetes. 2009;58:A30. Chem. 2009;285(9):6477-6488. between necrotic and apoptotic cell death by the ability to induce sterile inflammation. Proc Natl 88. Carmi Y, Voronov E, Dotan S, et al. The role of 53. Andrei C, Margiocco P, Poggi A, Lotti LV, Torrisi MR, Acad Sci U S A. 2010;107(6):2574-2579. macrophage-derived IL-1 in induction and main- Rubartelli A. Phospholipases C and A2 control tenance of angiogenesis. J Immunol. 2009;183: 71. Pollock AS, Turck J, Lovett DH. The prodomain of lysosome-mediated IL-1 beta secretion: implications 4705-4714. for inflammatory processes. Proc Natl Acad Sci interleukin 1alpha interacts with elements of the 89. Coxon A, Bolon B, Estrada J, et al. Inhibition of USA.2004;101(26):9745-9750. RNA processing apparatus and induces apopto- sis in malignant cells. FASEB J. 2003;17(2):203- interleukin-1 but not tumor necrosis factor sup- 54. Perregaux DG, McNiff P, Laliberte R, Conklyn M, 213. presses neovascularization in rat models of cor- Gabel CA. ATP acts as an agonist to promote neal angiogenesis and adjuvant arthritis. Arthritis 72. Abbate A, Salloum FN, Vecile E, et al. Anakinra, a stimulus-induced secretion of IL-1 beta and IL-18 Rheum. 2002;46(10):2604-2612. recombinant human interleukin-1 receptor antag- in human blood. J Immunol. 2000;165(8):4615- 90. Voronov E, Shouval DS, Krelin Y, et al. IL-1 is re- 4623. onist, inhibits apoptosis in experimental acute myocardial infarction. Circulation. 2008;117(20): quired for tumor invasiveness and angiogenesis. 55. Solle M, Labasi J, Perregaux DG, et al. Altered 2670-2683. Proc Natl Acad Sci U S A. 2003;100(5):2645- cytokine production in mice lacking P2X(7) recep- 2650. 73. Di Paolo NC, Miao EA, Iwakura Y, et al. Virus tors. J Biol Chem. 2001;276(1):125-132. 91. Olson JL, Courtney RJ, Rouhani B, Mandava N, binding to a plasma membrane receptor triggers Dinarello CA. Intravitreal anakinra inhibits choroi- 56. Agostini L, Martinon F, Burns K, McDermott MF, interleukin-1 alpha-mediated proinflammatory ␤ dal neovascular membrane growth in a rat model. Hawkins PN, Tschopp J. NALP3 forms an IL-1 macrophage response in vivo. Immunity. 2009; Ocul Immunol Inflamm. 2009;17(3):195-200. processing inflammasome with increased activity 31(1):110-121. in Muckle-Wells auto-inflammatory disorder. Im- 92. Antin JH, Weisdorf D, Neuberg D, et al. Interleukin-1 74. Kelk P, Claesson R, Chen C, Sjostedt A, munity. 2004;20(3):319-325. blockade does not prevent acute graft-versus- Johansson A. IL-1beta secretion induced by host disease: results of a randomized, double- 57. Hoffman HM, Mueller JL, Broide DH, Wanderer AA, Aggregatibacter (Actinobacillus) actinomyce- blind, placebo-controlled trial of interleukin-1 re- Kolodner RD. Mutation of a new gene encoding a temcomitans is mainly caused by the leuko- ceptor antagonist in allogeneic bone marrow putative pyrin-like protein causes familial cold autoin- toxin. Int J Med Microbiol. 298(5-6):529-541, transplantation. Blood. 2002;100(10):3479-3482. flammatory syndrome and Muckle-Wells syndrome. 2008. Nat Genet. 2001;29(3):301-305. 93. Antin JH, Weinstein HJ, Guinan EC, et al. Re- 75. Marrack P, McKee AS, Munks MW. Towards an combinant human interleukin-1 receptor antago- 58. Bergsbaken T, Fink SL, Cookson BT. Pyroptosis: understanding of the adjuvant action of alu- nist in the treatment of steroid-resistant graft- host cell death and inflammation. Nat Rev Micro- minium. Nat Rev Immunol. 2009;9(4):287-293. biol. 2009;7(2):99-109. versus-host disease. Blood. 1994;84(4):1342- 76. Ben-Sasson SZ, Hu-Li J, Quiel J, et al. IL-1 acts 1348. 59. Dinarello CA, Ikejima T, Warner SJ, et al. Interleu- directly on CD4 T cells to enhance their antigen- 94. Wood DD, Ihrie EJ, Dinarello CA, Cohen PL. Iso- kin 1 induces interleukin 1. I. Induction of circulat- driven expansion and differentiation. Proc Natl lation of an interleukin-1-like factor from human ing interleukin 1 in rabbits in vivo and in human Acad Sci U S A. 2009;106(17):7119-7124. joint effusions. Arthritis Rheum. 1983;26(8):975- mononuclear cells in vitro. J Immunol. 1987; 77. van de Veerdonk FL, TeirlinckAC, Kleinnijenhuis J, 983. 139(6):1902-1910. et al. Mycobacterium tuberculosis induces IL-17A 95. Duff GW, Atkins E, Malawista SE. The fever of 60. Fantuzzi G, Ku G, Harding MW, et al. Response responses through TLR4 and dectin-1 and is critically gout: urate crystals activate endogenous pyrogen to local inflammation of IL-1beta converting en- dependent on endogenous IL-1. J Leukoc Biol. 2010; production from human and rabbit mononuclear zyme-deficient mice. J Immunol. 1997;158(4): 88(2):227-232. phagocytes. Trans Assoc Am Physicians. 1983; 1818-1824. 78. Joosten LA. Excessive interleukin-1 signaling de- 96:234-245. 61. Joosten LA, Netea MG, Fantuzzi G, et al. Inflam- termines the development of Th1 and Th17 re- 96. So A, De Smedt T, Revaz S, Tschopp J. A pilot matory arthritis in caspase 1 gene-deficient mice: sponses in chronic inflammation. Arthritis Rheum. study of IL-1 inhibition by anakinra in acute gout. contribution of proteinase 3 to caspase 1- 2010;62(2):320-322. Arthritis Res Ther. 2007;9(2):R28. independent production of bioactive interleukin- 79. Chung Y, Chang SH, Martinez GJ, et al. Critical 97. Terkeltaub R. Update on gout: new therapeutic 1beta. Arthritis Rheum. 2009;60(12):3651-3662. regulation of early Th17 cell differentiation by strategies and options. Nat Rev Rheumatol. 62. Guma M, Ronacher L, Liu-Bryan R, Takai S, interleukin-1 signaling. Immunity. 2009;30(4):576- 2010;6(1):30-38. Karin M, Corr M. Caspase 1-independent activation 587. 98. So A, De Meulemeester M, Pikhlak A, et al. Ca- of interleukin-1beta in neutrophil-predominant inflam- 80. Koenders MI, Devesa I, Marijnissen RJ, et al. nakinumab for the treatment of acute flares in mation. Arthritis Rheum. 2009;60(12):3642-3650. Interleukin-1 drives pathogenic Th17 cells during difficult-to-treat gouty arthritis: Results of a multi- 63. Zhou R, Tardivel A, Thorens B, Choi I, Tschopp J. spontaneous arthritis in interleukin-1 receptor an- center, phase II, dose-ranging study. Arthritis Thioredoxin-interacting protein links oxidative tagonist-deficient mice. Arthritis Rheum. 2008; Rheum. 2010;62(10):3064-3076. stress to inflammasome activation. Nat Immunol. 58(11):3461-3470. 99. Martinon F, Mayor A, Tschopp J. The inflam- 2010;11(2):136-140. 81. Nakae S, Saijo S, Horai R, Sudo K, Mori S, masomes: guardians of the body. Annu Rev Im- 64. Meissner F, Molawi K, Zychlinsky A. Superoxide Iwakura Y. IL-17 production from activated T cells munol. 2009;27:229-265. From www.bloodjournal.org by guest on September 2, 2017. For personal use only.

3732 DINARELLO BLOOD, 7 APRIL 2011 ⅐ VOLUME 117, NUMBER 14

100. Joosten LA, Netea MG, Mylona E, et al. Engage- Blockade of interleukin 1 in type 1 diabetes melli- controls adipocyte differentiation and insulin sen- ment of fatty acids with Toll-like receptor 2 drives tus. Nat Rev Endocrinol. 2010;6(3):158-166. sitivity. Cell Metab. 2010;12(6):593-605. interleukin-1beta production via the ASC/caspase 105. Maedler K, Sergeev P, Ris F, et al. Glucose- 109. Joosten LA, Smeets RL, Koenders MI, et al. Inter- 1 pathway in monosodium urate monohydrate induced beta cell production of IL-1beta contrib- leukin-18 promotes joint inflammation and in- crystal-induced gouty arthritis. Arthritis Rheum. utes to glucotoxicity in human pancreatic islets. 2010;62(11):3237-3248. duces interleukin-1-driven cartilage destruction. J Clin Invest. 2002;110(6):851-860. Am J Pathol. 2004;165(3):959-967. 101. Dinarello CA. How interleukin-1beta induces gouty arthritis. Arthritis Rheum. 2010;62(11):3140-3144. 106. Dinarello CA, Donath MY, Mandrup-Poulsen T. 110. Chevalier X, Goupille P, Beaulieu AD, et al. Intra- Role of IL-1beta in type 2 diabetes. Curr Opin En- 102. Ghezzi P, Dinarello CA, Bianchi M, Rosandich ME, articular injection of anakinra in osteoarthritis of docrinol Diabetes Obes. 2010;17(4):314-321. Repine JE, White CW. Hypoxia increases production the knee: a multicenter, randomized, double- of interleukin-1 and tumor necrosis factor by human 107. Masters SL, Dunne A, Subramanian SL, et al. blind, placebo-controlled study. Arthritis Rheum. mononuclear cells. Cytokine. 1991;3(3):189-194. Activation of the NLRP3 inflammasome by islet 2009;61(3):344-352. 103. Emsley HC, Smith CJ, Georgiou RF, et al. A ran- amyloid polypeptide provides a mechanism for 111. Abbate A, Kontos MC, Grizzard JD, et al. domised phase II study of interleukin-1 receptor enhanced IL-1beta in type 2 diabetes. Nat Immu- Interleukin-1 blockade with anakinra to prevent antagonist in acute stroke patients. J Neurol Neu- nol. 2010;11(10):897-904. adverse cardiac remodeling after acute myo- rosurg Psychiatry. 2005;76(10):1366-1372. 108. Stienstra R, Joosten LA, Koenen T, et al. The cardial infarction. Am J Cardiol. 2010;105(10): 104. Mandrup-Poulsen T, Pickersgill L, Donath MY. inflammasome-mediated caspase-1 activation 1371-1377. From www.bloodjournal.org by guest on September 2, 2017. For personal use only.

2011 117: 3720-3732 doi:10.1182/blood-2010-07-273417 originally published online February 8, 2011

Interleukin-1 in the pathogenesis and treatment of inflammatory diseases

Charles A. Dinarello

Updated information and services can be found at: http://www.bloodjournal.org/content/117/14/3720.full.html Articles on similar topics can be found in the following Blood collections Free Research Articles (4676 articles) Immunobiology and Immunotherapy (5507 articles) Review Articles (713 articles)

Information about reproducing this article in parts or in its entirety may be found online at: http://www.bloodjournal.org/site/misc/rights.xhtml#repub_requests

Information about ordering reprints may be found online at: http://www.bloodjournal.org/site/misc/rights.xhtml#reprints

Information about subscriptions and ASH membership may be found online at: http://www.bloodjournal.org/site/subscriptions/index.xhtml

Blood (print ISSN 0006-4971, online ISSN 1528-0020), is published weekly by the American Society of Hematology, 2021 L St, NW, Suite 900, Washington DC 20036. Copyright 2011 by The American Society of Hematology; all rights reserved.